# nethical6/conversation-steganography

Use LLMs to hide messages inside normal looking conversations

Repository: https://github.com/nethical6/conversation-steganography
Canonical: https://ross.abutalabs.com/products/conversation-steganography
Language: Go
License: GPL-3.0
License Family: copyleft
Topics: llm-tools, steganography
Last push: 2026-07-18T13:20:49+00:00

## Health v2 (maintenance only)
Score: 55/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 93, release rhythm 35, longevity 3
- inputs: {"age_days": 47, "days_push": 46, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, young
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1239, forks 89 (observed 2026-08-28T04:04:06.104555+00:00)

## What it is
A Go CLI tool that hides encrypted secret messages inside natural-looking chat text generated by a local LLM (GPT-2), enabling covert communication over any messaging app. It is an educational proof of concept that encrypts messages locally and disguises them as innocent cover text that a recipient can decode.

## Use cases
- hide secret messages inside normal-looking chat conversations
- send covert messages over WhatsApp or Telegram without detection
- exchange encrypted messages that look like innocent small talk
- experiment with LLM-based text steganography
- simulate two-party covert communication locally
- research steganography detection of AI-generated cover text

## When to choose
- you need plausible-deniability messaging that doesn't look encrypted to observers
- you want a local, offline steganography tool with no cloud dependency
- you're researching or learning about LLM-based steganography techniques

## When to avoid
- you need production-grade security - it's a proof of concept with known weaknesses
- you require strong protection against steganalysis, which is actively being developed
- you want a polished GUI or mobile app rather than a CLI
- your use case involves any illegal activity - it's for educational and research purposes only

## Facets
- artifact type: cli-tool
- maturity: experimental
- function: cryptography, nlp, llm-inference, security, cli
- domain: privacy, security, large-language-models, censorship-circumvention
- platform: windows, cli, cross-platform
- tags: steganography, llm-steganography, covert-communication, gpt-2, proof-of-concept, encrypted-messaging, natural-language-processing, linux, macos

## Member repositories
- nethical6/conversation-steganography (main) score 55

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:06.104555+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T05:09:58.389356+00:00, confidence not recorded.
  - readme: https://github.com/nethical6/conversation-steganography (fetched 2026-08-28T04:04:06.104555+00:00, sha 6f01aaca066b)
- Data as of 2026-08-30T08:39:29.467469+00:00.